Online research could help anticipate waves of Covid-19

How can one best anticipate the emergence of a new wave of epidemics? This question has been asked of the leaders of most countries in the world for over a year. Indeed, as long as collective immunity has not been achieved, there is always a risk of relapse. To solve this problem, New York University researchers decided to use Google Trends to study online search data in the United States.

They then decided to differentiate human behavior into two categories. The first, called the “mobility index,” relates to social activities that are at risk of infection, such as walking around the world. B. Requests to find a cinema nearby, to buy plane tickets or to find out about the closing times of a bar.

Watching social networks can also be rich in lessons

On the contrary, the second, called the “isolation index”, consists of practices that are unlikely to pose a contamination risk. This is the case with requests to exercise at home or with people who want to be delivered to their home.

These indicators were examined in the period from March to June 2020, i. H. At the heart of the first wave. Overall, the countries with the highest mobility index recorded the highest number of Covid cases during this period. Conversely, a decrease in contamination was observed when research related to travel decreased.

If this data can be useful for the rest of the pandemic, the researchers also believe that this technique could be based on artificial intelligence to “better anticipate other future epidemics”. “”

You are not the first to come to this conclusion. To name just one example: Scientists from the IMT School for Advanced Studies in Italy found that signals were already visible on Twitter in December 2019 that indicated the emergence of the first cases of Covid-19.

Back to top button